I agree to a point, but different materials are more/less tolerant of manufacturing variances and also have different fatigue/failure properties.
I honestly don’t know whether carbon cranks fail less often compared to high end aluminum cranks, but that’s where I’d place my bet if I had to.
I’ve seen several aluminum crank arms snapped off and I’ve never seen a carbon crank fail catastrophically.
I’d also speculate on whether these failures are really a manufacturing/design defect or caused by damage. Carbon vs. aluminum have very different properties that would cause different failure patterns. A bad pedal strike on an aluminum crank might cause a small crack, but very unlikely it would rip a crank arm off on a single strike. Once you have a crack in an aluminum part, it’s a ticking time bomb. You might ride the crank for another day or 3 more years, but it will eventually fail once it’s cracked. You could have something similar with a carbon crank, but much less likely based on the material properties. More likely for a carbon crank to fail at the time of impact vs. delayed (unpredictable) failure.
The thing is these failures appear to be limited to the 2 piece bonded crank arms. This includes the right side Ultegra and Dura Ace crank arms, and the left side Dura Ace crank arms. The bonding fails for whatever reason and the two halves split apart very cleanly. The aluminum cracks and breaks because it is now under a load it was never designed to handle.
This sounds similar to how some older bonded carbon forks fail. Neither the crown nor the leg materials fracture. Instead they separate rather cleanly at the joint.
Apparently so, I’m sure it was GCN who reported it at the time as I was following his progress on there.
The crank snapped close to the pedal spindle and it was stated that his broken elbow had caused an imbalance due to the way he was holding himself on the bike.
Whether they could see that from power data I don’t know!
Found the tweet, assume it was a Dura Ace left side crank. You can see the clean split from the bonded halves. I don’t buy for a second that this was from a power imbalance, especially if that explanation is coming from GCN.
The pictures I’ve seen of these failures don’t point to a material properties issue, but rather to a bonded multi-part assembly failing at the bond. In other words - could be made of aluminum, carbon, wood or steel, if the bond fails, then the individual parts will follow and fail under the now-abnormal load.
There are potential manufacturing issues with such a design (incorrect mix, incorrect curing, name it), and also potential long-term reliability issues (resistance to moisture, for example). Of course mechanical damage from hits and scrapes can amplify the problems.
Looks like bond separation followed by fatigue failure. Would need to see a close-up of both pieces to get a better idea. Fatigue failures display pretty recognizable patterns in the failure surface. But there’s no way this can be blamed on a pedaling imbalance.
The problemo is that there are probably 1000X as many aluminum cranks out on the market as there are carbon cranks. Not even exaggerating on that number because with the exception of welded 4130 chromoly BMX cranks, the vast majority of cranks on bikes between $300 and $5000 are aluminum. So naturally you’re going to see higher failure numbers.
Shimano’s “hollow forged” crankarms are a decades-old design that saw a blip in their otherwise stellar track record. They should replace OP’s 6800 cranks with the 8000 series equivalent.
This is called ‘rationalization’. I know a guy who pedals around with only one leg. His cranks never break. Or, they haven’t yet. I don’t think Mark was more imbalanced than that.
Your point about exposing the cranks to humid conditions is valid, for sure. I bet also if you get them wet and then ride them around in freezing weather it’s a big issue as well. I’ve known people to try various tricks like clear nail polish or silicon sealant along the seams to keep the water out. That’s intuitive…don’t know if it works.
Couple points…I’m not willing to spend >$100 on equipment that’s good for ~two years…and, second, are there any other brand cranks in your area that failed? I’m at 7, I think, not counting non-drive-side crank arms that fall off. A lot of times the non-drive-side crank arm falling off is related to user error. Drive side crank spider blowing apart is not user error.
My own personal bias is that, unless you smash it on a rock or a curb, no crank should fail ever. Except for the bearings. I’d never heard of a crank failing before Shimano hollowtech. Hollowtech is incredible technology but those cranks have a fatal design flaw.
Yeah, I get that there is probably a population bias here, but my point was focused on higher end cranks. In that world, I think the dominant market players are SRAM Force/Red (carbon) vs. Shimano Ultegra/DA (aluminum). I don’t know what the global mix is between those, but it’s not that dramatic. Maybe 3:1 in favor of shimano these days? 8 years ago (before di2 became so big), SRAM owned the road market around here and it was probably 3:1 in favor of SRAM. I am just throwing out personal observations, so I still don’t know if it’s statistically meaningful, but I don’t think it’s fair to compare all the low end aluminum stuff to carbon cranks since a carbon crank is typically high end. Bike parts are a little unique in that the higher end (lighter) stuff is typically more prone to failure than the cheaper stuff.
I honestly don’t know what technique they use to manufacture them, the marketing terms are not helpful as they’re all “Hollowtech II” now.
This video has some closeups of the backside of the 5800, 6800 and r8000 if you’re interested. You can see the seam where the two halves meet on the Ultegra cranks, but that’s not the case on the 105. Maybe they weld it together and grind it smooth?
Cycling components are tested against a number of QA tests. That means pure force force and long term use - an ultra endurance athlete may not be pushing huge watts but consider how long the components are under strain and some of the conditions they are in.
I know a guy who pedals around with only one leg. His cranks never break. Or, they haven’t yet. I don’t think Mark was more imbalanced than that.